Hip fractures are an important cause of mortality and functional dependence among older adults and approximately 340,000 hip fractures occur annually in this country. The proposed project brings together 3 cohorts of 2819 patients with hip fracture to create the largest dataset of hip fracture patients with follow-up information on functional outcomes. This dataset will be used to address 2 important topics in hip fracture research: a) the characterization of patient subgroups that could be targeted for future clinical interventions, and b) the role of early surgical intervention. The specific aims are: a) to combine the Mount Sinai, Baltimore, and Hospital for Joint Disease hip fracture cohorts to produce a dataset with information on patient characteristics, pre-fracture function, process of inpatient care, and longitudinal follow-up on function and other outcomes; b) to better characterize the mortality and functional outcomes of subgroups of hip fracture patients; c) to test how pre-fracture function, patient characteristics, and co-morbid conditions affect functional outcomes for patients as a whole and for major subgroups; and d) to test the impact of early surgery for patients with hip fracture. Our research methods include: a) assembling the dataset and creating and testing common definitions of study variables from the 3 cohorts; b) using cluster analysis to identify clinically important subgroups and to examine differences in mortality and functional outcomes among the subgroups; and c) analyzing the determinants of mortality and functional outcome, including the effect of early surgery. In the course of doing this, we will extend the methodological work on a) how mortality can be incorporated into analyses of functional outcome and b) how observational data can be analyzed to assess the effectiveness of interventions that are not likely to be evaluated in randomized trials. These 2 latter topics are critical issues to a series of other clinical topics in aging. The methodologies we propose to develop and test has wide applicability to other clinical problems of aging where functional status is an important outcome and mortality is also high.